REVIEW Open Access

Delayed initiation of antenatal care andassociated factors in Ethiopia: a systematicreview and meta-analysisGezahegn Tesfaye1,2* , Deborah Loxton1, Catherine Chojenta1, Agumasie Semahegn2 and Roger Smith3

Abstract

Background: Antenatal care uptake is among the key indicators for monitoring the progress of maternal outcomes.Early initiation of antenatal care facilitates the timely management and treatment of pregnancy complications toreduce maternal deaths. In Ethiopia, antenatal care utilization is generally low, and delayed initiation of care is verycommon. We aimed to systematically identify and synthesize available evidence on delayed initiation of antenatalcare and the associated factors in Ethiopia.

Methods: Studies published in English from 1 January 2002 to 30 April 2017 were systematically searched fromPubMed, Medline, EMBASE, CINAHL and other relevant sources. Two authors independently reviewed the identifiedstudies against the eligibility criteria. The included studies were critically appraised using the Joanna Briggs-MAStARIinstrument for observational studies. Meta-analysis was conducted in RevMan v5.3 for Windows using a Mantel–Haenszelrandom effects model. The presence of statistical heterogeneity was checked using the Cochran Q test, and its level wasquantified using the I2 statistics. Pooled estimate of the proportion of the outcome variable was calculated. Pooled OddRatios with 95% CI were calculated to measure the effect sizes.

Result: The pooled magnitude of delayed antenatal care in Ethiopia was 64% (95% CI: 57%, 70%). Maternal age (OR = 0.70; 95% CI: 0.53, 0.93), place of residence (OR = 0.29, 95% CI: 0.16, 0.50), maternal education (OR = 0.49; 95% CI: 0.38, 0.63),husband’s education (OR = 0.44; 95% CI: 0.23, 0.85), maternal occupation (OR = 0.75; 95% CI: 0.61, 0.93), monthly income(OR = 2.06; 95% CI: 1.23, 3.45), pregnancy intention (OR = 0.49; 95% CI: 0.40, 0.60), parity (OR = 0.46; 95% CI: 0.36, 0.58),knowledge of antenatal care (OR = 0.40; 95% CI: 0.32, 0.51), women’s autonomy (OR = 0.38; 95% CI: 0.15, 0.94), partnerinvolvement (OR = 0.24; 95% CI: 0.07, 0.75), pregnancy complications (OR = 0.23; 95% CI: 0.06, 0.95), and means ofidentifying pregnancy (OR = 0.50; 95% CI: 0.36, 0.69) were significantly associated with delayed antenatal care.

Conclusion: Improving female education and women’s empowerment through economic reforms, strengthening familyplanning programs to reduce unintended pregnancy and promoting partner involvement in pregnancy care couldreduce the very high magnitude of delayed antenatal care in Ethiopia.

Trial registration: CRD42017064585.

Keywords: Delayed antenatal care, Associated factors, Ethiopia, Systematic review, Meta-analysis

* Correspondence: gezites@gmail.com; gezahegntesfaye.girma@uon.edu.au1Research Centre for Generational Health and Ageing, Faculty of Health andMedicine, University of Newcastle, Newcastle, Australia2School of Public Health, College of Health and Medical Sciences, HaramayaUniversity, Harar, EthiopiaFull list of author information is available at the end of the article

© The Author(s). 2017 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, andreproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link tothe Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

Tesfaye et al. Reproductive Health (2017) 14:150 DOI 10.1186/s12978-017-0412-4

Plain English summaryA professional care provided to women during preg-nancy is called antenatal care. Antenatal care plays agreat role in the improvement of maternal health. InEthiopia and other sub-Saharan African countries, ante-natal care utilization among pregnant women was low.Moreover, the pregnant women in Ethiopia and otherdeveloping countries tend to postpone their first ante-natal care clinic visit into the later months of pregnancy.This study summarized the existing evidence on thelevel of late antenatal care visit and its contributing fac-tors among pregnant women in Ethiopia. Using differentdatabases and other sources, this review identifiedtwenty two relevant studies that reported late antenatalcare clinic visit and its influencing factors among preg-nant women in Ethiopia. We summarized and analysedthe reports from the twenty two studies and put a com-bined assessment result on the level of late antenatalcare and associated factors. Based on our review, nearlytwo thirds of the pregnant women in Ethiopia madetheir first antenatal care clinic visit late in their preg-nancy. Women’s non-attendance of education, husband’snon-attendance of education, women’s older age, ruraldwelling, having previous births, unintended pregnancy,women’s unemployment, low monthly income, lack ofknowledge about antenatal care, lack of women’s decisionmaking power, no partner involvement, and not facingproblems during pregnancy were factors associated withhigher level of women’s late appearance to antenatal care.Nationwide all rounded efforts targeting the major contrib-uting factors should be established to alleviate women’s lateantenatal care utilization in the country.

BackgroundThe burden of maternal mortality remains hugely variedbetween developing and developed countries [1]. In de-veloping countries, the overall life time risk of woman’sdeath due to pregnancy and related causes is estimatedto be 1 in 180, while for developed countries it is about1 in 4900 [2]. The maternal mortality ratio in Ethiopia isstill high at 353 per 100,000 live births in 2015 [3], andit remains among the highest in the world. In developingcountries like Ethiopia, obstetric complications duringpregnancy and childbirth are the leading causes of deathamong reproductive aged women [3, 4]. It is generallyrecognized that a lack of access to, and inadequateutilization of, antenatal care (ANC) during pregnancycontributes to adverse maternal health outcomes such asmaternal mortality [5, 6], something which is more com-mon in resource-poor settings. Antenatal care uptake isone of the key indicators for monitoring the progress ofimproving maternal outcomes. Early initiation of ANCfacilitates the timely management and treatment of preg-nancy complications to reduce maternal deaths [7].

In Ethiopia, the main direct causes of maternal mortal-ity are haemorrhage, hypertensive disorders of preg-nancy, unsafe abortion and puerperal sepsis [8, 9]. Thesecomplications can be averted or otherwise treatedthrough providing skilled care during pregnancy, childbirth and in the postnatal period [2]. In 2002, the WorldHealth Organization (WHO) recommended that preg-nant women make at least four ANC visits [10]; in 2016this recommendation was modified to at least eight visits[11], with the first ANC visit to be undertaken before the12th week of pregnancy. While there has been markedprogress in the uptake of at least one ANC attendance inEthiopia [12–17], there has been suboptimal attendance ofthe recommended visits [4, 13, 18, 19]. Of even more con-cern was the substantial proportion of women who de-layed their first ANC visit to the second or third trimesterof pregnancy [5, 19–21]. According to the NationalDemographic and Health Survey Report of Ethiopia [20],in 2014 more than three quarters of pregnant women ini-tiated their first visit after 16 weeks of pregnancy. Earlyinitiation of ANC plays a paramount role in enhancingmaternal health as it provides an opportunity for the earlyscreening, treatment and referral of pregnancy complica-tions [11]. Evidence has shown that pregnant women whoinitiate ANC early were less likely to develop unfavourableobstetric outcomes as compared to women who enteredinto care after the first trimester [22, 23].The key challenges that women face when seeking ma-

ternal health services were clearly explained in the threedelays model [24]. This model described the barriers toutilizing maternal health services at three interrelatedlevels before the occurrence of maternal death. At thefirst level, the home or community level, women may bedelayed from seeking ANC due to factors such as thelow social status of women in relation to decision-making, poor awareness of pregnancy or birth complica-tions, previous poor experience of care, traditional orsocial practices during pregnancy or childbirth, acceptanceof maternal death as normal and financial dependency. InEthiopia, there is huge gap in the level of income amongwomen and men especially in rural parts of Ethiopia, andwomen are less empowered to access and control house-hold resources [25]. This could influence their capacity tomake decisions about utilization of maternal care. More-over, the financial burden associated transportation to andfrom the facility and the costs incurred for the maternalcare itself profoundly diminished the uptake of the care[26]. In the second level, there may be a delay in reachinga health facility which might be due to distance, unavail-ability of infrastructure (road or transportation) or difficultterrain. The third level of delay (delay in receiving ad-equate care) might be related to a shortage of, or inad-equately trained health staff, and unavailability of medicalsupplies and equipment.

Tesfaye et al. Reproductive Health (2017) 14:150 Page 2 of 17

Several studies [27–33] have investigated factors af-fecting delayed attendance of ANC in Ethiopia. None-theless, none of these studies have systematicallyreviewed the factors to show their overall pooled effecton delayed initiation of ANC at the national level. Inaddition, there were inconsistencies in attributing theinfluence of the factors on late initiation of ANC acrossvarious studies. For instance, there were incongruent find-ings on the influence of maternal education [34–37], ma-ternal age [32–34, 36], place of residence [28, 32, 38, 39],maternal occupation [30, 34, 37, 40], marital status[32, 36, 37], husband’s education [31, 32, 41], previousexperience of using ANC [32, 33, 35] and history ofabortion [31, 32, 42] on delayed initiation of ANCamong many other factors. Hence, demonstrating apooled effect of the factors on delayed initiation ofANC was warranted.Previous systematic reviews conducted in developing

[43, 44] and developed [45] countries have mainlyreviewed evidence on the adequacy of the utilization ofANC and its related factors. In particular, the reviewscovered larger geographical regions and hence failed toreflect country specific situations. Moreover, these re-views did not centre on delayed initiation of ANC as aprimary outcome of interest. The objective of this reviewis to systematically identify and synthesize existing evi-dence to understand the level of delayed initiation ofANC and associated factors among reproductive agedwomen in Ethiopia.

MethodDevelopment of the review methodThe methodology of this systematic review was developedbased on the Preferred Reporting Items for SystematicReviews and Meta-Analyses Protocols (PRISMA-P) 2015Statement [46] and the items in the PRISMA-P checklistwere addressed (Additional file 1). The four phases thatwere drawn from the PRISMA flow chart ( [47]) were doc-umented in the results to show the study selection processfrom initially identified records to finally included studies.The protocol for this systematic review and meta-analysiswas registered in international prospective register of sys-tematic reviews (PROSPERO) and obtained the registra-tion number (CRD42017064585).

Search strategyThe literature search was carried out by the primary au-thor (GT). The search was limited to papers publishedin English from 1 January 2002 to 30 April 2017. Theyear 2002 was selected, since WHO had introduced theFocused ANC model [10] by this year. We appliedMeSH terms, Emtree, CINAHL headings and combinedkey words to identify studies in the databases. Majormedical electronic databases such as PubMed, Medline

(OVID interface), Excerpta Medica (Embase) (OVIDinterface), and CINAHL (EBSCO host) were used toidentify relevant literature for the review. To cover greyliterature, we hand-searched literature using the Googlesearch engine and Google Scholar; official WHO websites;online libraries of academic and government institutionsand references of electronically identified articles. Thesearch strings or terms were stemmed from the followingkey words: delayed initiation, ANC, associated factors,and Ethiopia. The search terms were used to retrieverelevant literature in combined form adapted to therequirement of the specific database. Further informa-tion regarding the search strategy of the selected da-tabases is attached (Additional file 2).

Eligibility criteriaWe included all observational studies as well as Demo-graphic and Health Surveys (DHS) reports. We consideredstudies that examined the level and factors associated withdelayed initiation of ANC among reproductive agedwomen (15–49 years) who were pregnant or gave birth atleast once and who live in Ethiopia. We included studiesthat defined the main outcome variable “delayed initiationof ANC” as entry into care after at least 12 weeks of preg-nancy, including studies that defined delayed initiation ofANC as entry into the care after 16 weeks of gestation.Studies that had been conducted in either a community orfacility setting and which involved analysis of primary orsecondary data were included. We included studies thathad measure of association statistics or had test statisticthat explicitly demonstrated the influence of the predic-tors on delayed initiation of ANC or had a crosstab show-ing the difference in magnitude of the outcome variable inthe categories of the predicting variables. We excluded re-views, editorials, case series and case reports on delayedinitiation of ANC. We also excluded studies that only re-ported qualitative findings on delayed ANC initiation. Instudies that reported both quantitative and qualitative re-sults, we only considered the quantitative findings.

Study selection procedureScreeningFirst studies were identified through applying the searchstrings and the filters in the databases as well as otherrelevant sources. The identified studies were exported tothe citation manager (EndNote) [48] and duplicates wereexcluded. The two authors (GT and AS) independentlyscreened the studies based on the information containedin the titles and abstracts according to the inclusion cri-teria. Based on this screening, the titles and abstracts ofthe studies were classified as included, excluded, andundecided. We then obtained the full texts of all theincluded and the undecided studies for further eligi-bility assessment.

Tesfaye et al. Reproductive Health (2017) 14:150 Page 3 of 17

Eligibility of studiesThe two authors (GT and AS) independently reviewed thefull texts of the included and undecided categories of thestudies against the eligibility criteria for final inclusion.Studies that were not eligible based on the examin-ation of the full-text were excluded and the reasonsfor the exclusion were described. Disagreements be-tween the two reviewers were resolved through dis-cussion and consensus.

Quality assessmentAll of the included studies were critically appraised fortheir validity. The two authors (GT and AS) checked themethodological robustness and validity of the findingsusing the JBI (Joanna Briggs Institute) Meta-Analysis ofStatistics Assessment and Review Instrument (MAStARI)[49]. Particular attention was given to a clear statement ofthe objective of the study, inclusion criteria, randomnessof subject selection, identification of the study subjects,and preciseness of measurement of outcomes of interestand use of appropriate statistical analysis method, as wellas documentation of sources of bias or confounding. Un-certainties were resolved by joint discussion between thereviewers. The level of agreement between the two re-viewers was judged using the Cohen’s Kappa (K) coeffi-cient statistics. To calculate “K” a two by two contingencytable was constructed with “High” and “Low” categories ofquality assessment provided independently by the two re-viewers based on set of criteria. We obtained “K” value of(0.80), and thus the level of agreement was satisfactory. Inorder to minimize publication bias, we searched and in-cluded both published and unpublished literature. We ob-tained unpublished literatures (grey literatures) throughhand-searching of online libraries of academic institutions,government organizations, and agencies in addition tousing Google search engine and Google scholar. We alsocontacted an author to seek data that was not clearly re-ported in the article.

Data extraction processA structured data extraction template in the form ofsummary table was constructed for the data abstraction.The two authors (GT and AS) systematically used thedata extraction template to abstract data. The summarytable contained list of items pertaining to the study char-acteristics to concisely present all the included studies.The specific list of items included; study year, design ofthe study, study setting, sample size, study subjects, datacollection method, and study specific predicting factors.A quantitative data of cross-tabulation between the sub-ject’s characteristics (predicting factors) and the outcomevariable was also systematically abstracted. During thedata extraction of the exposure variables, we categorizedthe individual classifications shown for each variable in

the studies into two (exposed with the outcome andnon-exposed with the outcome). The non-exposed cat-egory was considered as the reference category of thevariables (e.g for place of residence, urban was the ex-posed and rural was the non-exposed category). Wethen put the corresponding combined numerical valueto make it ready for the quantitative synthesis. Duringthe data extraction, one of the papers (Bayou et al.2016) reported missing and incomplete data, and theprincipal author of the publication was contacted torequest further data via email. We received a re-sponse from the author, and were provided with therequested data. Disagreements between the two re-view authors were resolved by face to face discussionand reached a consensus.

Data synthesis and statistical analysisThe individual studies were concisely described using asummary table. The summary table particularly de-scribed the characteristics of the included studies andthe main findings. We conducted the quantitative syn-thesis using the Cochrane community Review ManagerSoftware (RevMan version 5.3 for windows) [50]. Sum-mary statistics (pooled effect sizes) in Odds Ratios with95% confidence intervals were calculated. We classifiedthe factors that showed significant association with theoutcome variable into three groups based on the threedelays model, though some overlapping exist betweenthem. Forest plots were used to graphically present themeta-analysis results. The presence of statistical hetero-geneity was checked by using the Chi2 test (Cochran Qtest) at p-value ≤0.05. The level of heterogeneity amongthe studies was quantified using the I2 statistics [51]where substantial heterogeneity was assumed if the I2

value was ≥50%. We conducted meta-analysis usingMantel–Haenszel random effects model when the studieswere substantially heterogeneous (I2 statistic ≥50%). Pooledestimate of the magnitude of the primary outcome variablewas conducted using stats direct (http://www.statsdirect.com) statistical software [52] using Stuart-Ord (inversedouble arcsine square root) method. We hypothesizedthat there could be variation in the factors that leadto delayed ANC between studies that defined delayedANC based on the WHO [10] recommendation with(≥12 weeks) and country specific recommendation[53] (≥16 weeks) due to the obvious difference inmagnitude of the outcome variable. Hence, subgroupanalysis was conducted based on comparison of outcomesfor studies that defined delayed initiation of ANC basedon (≥12 weeks) and (≥16 weeks), provided an adequatenumber of studies were available in the two groups. Theresult of the review was reported according to thePRISMA guideline for reporting [54].

Tesfaye et al. Reproductive Health (2017) 14:150 Page 4 of 17

ResultsDescription of the studiesWe retrieved 2975 studies through searching the majorhealth and medical electronic databases and other rele-vant sources. From all the identified studies, 1006 arti-cles were removed due to duplication while 1969 studieswere retained for further screening. The remaining 1969studies were then screened for their eligibility based onthe title and abstract. Accordingly, 1867 studies were ex-cluded because of the incompatibility of the content pre-sented in the title and abstract of the studies with ourreview topic. Hence, the full text of the remaining 102studies were assessed for eligibility. During the full text as-sessment, 80 studies were excluded from the review be-cause of duplication, inconsistent study outcome, orirrelevant target participants. The remaining twenty twostudies were critically appraised and included in the re-view. After the critical appraisal of the studies, we ex-cluded one study from the quantitative synthesis due tothe relatively poor methodological quality and inconsistentstatistical report. Finally, twenty one studies were includedfor the pooled estimation of delayed initiation of ANCand factor analysis (Fig. 1). Among the included studies,there were seventeen published articles, three master the-ses, and one Ethiopian Demographic and Health Survey(DHS data). All of the included studies were cross-sectional by design and seventeen of the studies wereconducted in a facility setting (Table 1). Ten of the studiesincluded in the quantitative synthesis reported delayed ini-tiation of ANC based on (≥ 12 weeks), and the remainingstudies reported it based on (≥ 16 weeks).With regards to the demographic characteristics, the

study participants in the included studies were pregnantwomen or women who have at least had one birth in thefive or three years prior to the studies. The age of theparticipants’ were ranged from 15 to 49 years. Large ma-jority of the participants in the included studies wereurban residents. Moreover, higher proportion of the par-ticipants in the included studies were married andattended formal education (primary school and above).

Magnitude of delayed initiation of ANCThe pooled estimate of the magnitude of delayed initi-ation of ANC in Ethiopia was 64% (95% CI: 57%, 70%)(Fig. 2). The result of the analysis for the magnitude ofdelayed initiation of ANC based on the studies that re-ported the outcome variable with (≥ 12 weeks) was 66%(95% CI: 56%, 76%), whereas based on the studies thatdefined the outcome variable with (≥ 16 weeks) it was62% (95% CI: 52%, 71%).

Factors associated with delayed initiation of ANCThe current review revealed various factors associatedwith delayed initiation of ANC in Ethiopia. Significantly

associated delay one factors include maternal age, mater-nal education, husband’s education, pregnancy intention,women’s autonomy, knowledge on ANC, partner involve-ment, pregnancy complication, and parity. Significantlyassociated delay two factors were maternal occupation,monthly income and place of residence. Means of checkingpregnancy was the only delay three factor that showed sta-tistically significant association with delayed ANC. The re-view also demonstrated that delay one factors such maritalstatus and history of abortion, and delay three factor (pre-vious use of ANC) were not significant predictors of de-layed attendance of ANC services (Table 2).

Maternal ageMaternal age was significantly associated with delayedinitiation of ANC. Women aged between 15 and 30 wereless likely to have delayed their first ANC booking ascompared to women aged 31 to 49 years of age (OR, 0.70;95% CI: 0.53, 0.93). However, the subgroup (delayed initi-ation ≥16 weeks) showed no association between maternalage and delayed booking of ANC (OR, 0.70; 95% CI:0.42, 1.19). But it did not affect the overall association.Random effect model was employed for the analysis as theI2 value was >50% (Fig. 3).

Maternal educationThe meta-analysis showed that maternal education wassignificantly associated with delayed ANC initiation. Theoverall Odds Ratio 0.49 at 95% CI: 0.38, 0.63 indicatedthat women who have attended primary or above levelof education were less likely to delay their first ANC visitas compared to women without formal education. Inspite of the heterogeneity of the studies, the findingshowed statistically significant association. The subgroupanalysis for studies with (≥12 weeks) (OR, 0.57; 95% CI:0.45, 0.72) and studies (≥16 weeks) (OR, 0.43; 95% CI:0.28, 0.67) both showed significant association betweenthe maternal educational status and delayed initiation ofANC. We used random effect model for the analysissince the I2 value was 75% (Fig. 4).

Place of residenceAccording to the factor analysis of the included studies,place of residence was significantly associated with de-layed initiation of ANC. Women who live in urban areawere less likely to have delayed initiation of ANC (OR,0.29, 95% CI: 0.16, 0.50). No difference was found interms of the direction of association between place ofresidence and delayed initiation of ANC in the sub-groups analysis. Random effect model was used for theanalysis since the heterogeneity test showed an overall I2

value of 89% (Fig. 5).

Tesfaye et al. Reproductive Health (2017) 14:150 Page 5 of 17

Pregnancy intentionThe review finding showed that women with intendedpregnancy were less likely to delay their ANC initiation(OR, 0.49; 95% CI: 0.40, 0.60). There was no differencebetween the subgroups in the direction of association.As the heterogeneity test indicated an I2 value of 59%,random effect was considered for the analysis (Fig. 6).Below are the descriptions of other factors that are as-

sociated with delayed initiation of ANC with the meta-analysis test statistics (Table 3).

Family monthly incomeMonthly average family income was significantly associ-ated with delayed ANC initiation. It was demonstratedthat there was increased odds of delayed initiation ofANC among women with an average family income of≤1000 ETB (50USD) compared to those women whoseaverage family income was >1000ETB (50USD) (OR,2.06; 95% CI: 1.23, 3.45). The association between familymonthly income and delayed initiation of antennal carewas not consistent across the two subgroups, where thesubgroup which defined the outcome variable based on

(≥16 weeks) showed insignificant association betweenmonthly income and delayed initiation of ANC (OR,2.26; 95% CI: 0.96, 5.29), whereas the subgroup (delayedANC ≥12 weeks) (OR, 1.77; 95% CI: 1.16, 2.72) showedsignificant association. Due to the heterogeneity ofthe studies (I2 = 91%), we used random effect modelfor the analysis.

Marital statusOur systematic review demonstrated that there was nosignificant association between marital status and de-layed initiation of ANC (OR, 0.81; 95% CI: 0.56, 1.16).The same was true in the subgroup analysis whichshowed no association between marital status and de-layed initiation of ANC. We assumed random effectmodel for the analysis since the I2 statistics showed pres-ence of heterogeneity (68%).

Maternal occupationThe overall Odds Ratio showed that there was signifi-cant association between maternal occupation and de-layed initiation of ANC (OR, 0.75; 95% CI: 0.61, 0.93).

Fig. 1 Schematic presentation of the PRISMA flow diagram to select and include studies

Tesfaye et al. Reproductive Health (2017) 14:150 Page 6 of 17

Table

1Descriptio

nof

thestud

ycharacteristicsfortheinclud

edstud

iesin

thereview

No

Autho

randyear

Settingof

thestud

yDesignof

thestud

ySample

size

Stud

ysubjects

Datacollectionmetho

dPrim

aryOutcomeof

Interest

Delayed

ANC

(definition

)Stud

yspecificpred

ictin

gfactorsfor

delayedup

take

ofANC

1Amen

tieet

al.2015

[39]

Com

mun

itybasedstud

yCross

sectional

536

Reprod

uctiveaged

wom

enwho

hadat

leaston

ebirthin

thefiveyearspriorto

thestud

y

Interviewer

administered

questio

nnaire

-Utilizationof

ANC

(uptake)

-Tim

ingof

firstANC

initiation

Entryin

tocare

after12

weeks

ofge

station

Placeof

reside

nce(living

inruralarea)

2Abo

sseet

al.2010

[27]

Com

mun

itybasedstud

yCross

sectional

710

Reprod

uctiveaged

wom

enwho

hadat

leaston

ebirthin

thefiveyearspriorto

thestud

y

Interviewer

administered

questio

nnaire

-Utilizationof

ANC

(uptake)

-Tim

ingof

firstANC

visit

Entryin

tocare

after12

weeks

ofge

station

Placeof

reside

nce(living

inruralarea)

3Abu

kaet

al.2014

[35]

Facilitybased

stud

yCross

sectional

406

Preg

nant

wom

enattend

ing

health

facility

Interviewer

administered

questio

nnaire

-Timingof

firstANC

booking

Entryin

tocare

after12

weeks

ofge

station

Age

(≥20

year),no

n-attend

ance

ofform

aled

ucation,

high

parity,

perceivedthat

timelyANCisno

tim

portant,no

thaving

inform

ation,

previous

non-useof

ANC

4Bayouet

al.2016

[36]

Com

mun

itybasedstud

yCross

sectional

814

Reprod

uctiveaged

wom

enwho

hadat

leaston

ebirthin

the

threeyearspriorto

thestud

y

Interviewer

administered

questio

nnaire

-Earlyinitiationof

ANC

-Atleastfour

ANC

visit

-Ade

quacyof

ANC

Entryin

tocare

after12

weeks

ofge

station

Uninten

dedpreg

nancyand

non-attend

ance

ofform

aled

ucation

5Belayneh

etal.2014

[34]

Facilitybased

stud

yCross

sectional

369

Preg

nant

wom

enattend

ing

ANCservicein

health

facility

Face-to-face

interview

techniqu

e-Tim

ingof

firstANC

booking

Entryin

tocare

after12

weeks

ofge

station

Non

-atten

danceof

form

aled

ucation,

olde

rage[30–49],previous

early

ANCvisit,pe

rceivedsufficien

tnu

mbe

rof

ANC(4+)

6Gud

ayu2015

[37]

Facilitybased

stud

yCross

sectional

390

Preg

nant

wom

enattend

ing

ANCservicein

health

facilities

Face-to-face

exitinterview

techniqu

e-LateANCbo

oking

Entryin

tocare

after12

weeks

ofge

station

Not

obtaininginform

ationon

right

timeto

initiate,pe

rceivedrig

httim

eto

book

ANC(12+

weeks),

non-autono

my,anduseof

urine

testto

iden

tifypreg

nancy

7Gud

ayuet

al.2014

[30]

Facilitybased

stud

yCross

sectional

407

Preg

nant

wom

enattend

ing

health

facility

Face-to-face

exitinterview

techniqu

e-Tim

ingof

firstANC

booking

Entryin

tocare

after12

weeks

ofge

station

Age

(>25),youn

gerageat

marriage

,preg

nancychecking

bymeans

othe

rthan

urinetest,p

erceived

right

time

tostartANC(12+

weeks),and

non-autono

my

8YilalaandSinishaw

2015

[33]

Facilitybased

stud

yCross

sectional

407

Preg

nant

wom

enattend

ing

antenatalcareclinicin

health

facility

Face-to-face

exitinterview

techniqu

e-Lateinitiationof

ANC

Entryin

tocare

after12

weeks

ofge

station

Non

-atten

danceof

form

aled

ucation,

poor

know

ledg

eof

ANC,n

otreceivingadvice

from

HEW

,not

gettingadvice

onANCbo

oking,

perceivedrig

httim

eof

ANC(12+

weeks)

9Ze

geye

etal.2013

[65]

Facilitybased

stud

yCross

sectional

446

Preg

nant

wom

enattend

ing

health

facility

Face-to-face

exitinterview

techniqu

e-EarlyANCvisit

Entryin

tocare

after12

weeks

ofge

station

Highparity,lack

ofknow

ledg

eof

ANC,

unintend

edpreg

nancy

10Tarikuet

al.2010

Facilitybased

stud

yCross

sectional

612

Preg

nant

wom

enattend

ing

health

facility

Face

toface

exitinterview

-Tim

ingof

firstANC

booking

Entryin

tocare

after12

weeks

ofge

station

Highparity,un

intend

edpreg

nancy,

obtainingadvice

onwhe

nto

book

firstANC

11CSA

2014

[20]

Com

mun

itybasedstud

yCross

sectional

(DHSdata)

1571

Reprod

uctiveaged

wom

enwho

hadat

leaston

ebirth

inthefiveyearspriorto

thesurvey

Interviewer

administered

questio

nnaire

-Tim

ingof

ANC

initiation

-Atleaston

eANC

visit

Entryin

tocare

after16

weeks

ofge

station

Placeof

reside

nce(living

inruralarea)

Tesfaye et al. Reproductive Health (2017) 14:150 Page 7 of 17

Table

1Descriptio

nof

thestud

ycharacteristicsfortheinclud

edstud

iesin

thereview

(Con

tinued)

No

Autho

randyear

Settingof

thestud

yDesignof

thestud

ySample

size

Stud

ysubjects

Datacollectionmetho

dPrim

aryOutcomeof

Interest

Delayed

ANC

(definition

)Stud

yspecificpred

ictin

gfactorsfor

delayedup

take

ofANC

12Dam

meet

al.2015

[28]

Facilitybased

stud

yCross

sectional

379

Preg

nant

wom

enattend

ing

ANCservicein

health

facilities

Face-to-face

exitinterview

techniqu

e-Tim

ingof

firstANC

booking

Entryin

tocare

after16

weeks

ofge

station

Non

-atten

danceof

form

aled

ucation,

ruralresiden

ce,low

income,having

noaw

aren

esson

timingof

ANC

13Ew

enetuet

al.2015

[29]

Facilitybased

stud

yCross

sectional

178

Preg

nant

wom

enattend

ing

ANCservicein

health

facility

Interviewer

administered

structured

questio

nnaire

Late

ANCinitiation

Entryin

tocare

after16

weeks

ofge

station

Non

-atten

danceof

education,

rural

reside

nce,no

historyof

prem

ature

birth,

late

recogn

ition

ofpreg

nancy,

andun

intend

edpreg

nancy

14Fisseh

aet

al.2015

[66]

Facilitybased

stud

yCross

sectional

410

Preg

nant

wom

enattend

ing

ANCservicein

health

facilities

Interviewer

administered

structured

questio

nnaire

Timingof

FirstANC

Booking

Entryin

tocare

after16

weeks

ofge

station

Nohistoryof

stillbirth,

nopreg

nancy

complications,lackof

know

ledg

eof

timeto

initiateANC,n

opartne

rinvolvem

enton

ANC

15Geb

remeskeletal.

2015

[40]

Facilitybased

stud

yCross

sectional

409

Preg

nant

wom

enattend

ing

ANCservicein

health

facility

Interviewer

administered

structured

questio

nnaire

Timingof

FirstANC

Atten

dance

Entryin

tocare

after16

weeks

ofge

station

Low

income,no

treceivingadvice

onwhe

nto

startANC,h

ouseho

ldfood

insecurity,un

intend

edpreg

nancy

16Giru

m2016

[38]

Facilitybased

stud

yCross

sectional

362

Preg

nant

wom

enattend

ing

ANCservicein

health

facilities

Face

toface

exitinterview

Timingof

FirstANC

Visit

Entryin

tocare

after16

weeks

ofge

station

Ruralresiden

ce,low

income,

non-attend

ance

ofed

ucation,

not

receivingadvice

ontim

ingof

visit

andun

intend

edpreg

nancy

17Gulem

aand

Berhane2017

[67]

Facilitybased

stud

yCross

sectional

960

Preg

nant

wom

envisitin

ghe

alth

facilitiesforthe

firsttim

e

Interviewer

administered

structured

questio

nnaire

Timingof

FirstANC

Visit

Entryin

tocare

after16

weeks

ofge

station

Une

mploymen

t,low

income,pe

rceived

ANCinitiationtim

e(16weeks

+),

unintend

edpreg

nancy,having

preg

nancycomplications

18Hailesilasieand

Enqu

selasie2010

[41]

Facilitybased

stud

yCross

sectional

419

Preg

nant

wom

enattend

ing

ANCat

governmen

the

alth

facilities

Face-to-face

interview

ofpreg

nant

wom

enLate

Initiationof

ANC

ServiceUtilization

Entryin

tocare

after16

weeks

ofge

station

Youn

gerage,no

n-attend

ance

ofform

aled

ucation,

low

perceivedbe

nefit

ofANC,uninten

dedpreg

nancy,pe

rceived

ANCinitiationtim

e(4-6

mon

ths)

19Hussenet

al.2016

[42]

Facilitybased

stud

yCross

sectional

255

Preg

nant

wom

enattend

ing

ANCat

governmen

the

alth

facilities

Interviewer

administered

structured

questio

nnaire

TimelyInitiationof

FirstANCVisit

Entryin

tocare

after16

weeks

ofge

station

Non

-atten

danceof

form

aled

ucation,

lack

ofknow

ledg

eof

ANC,late

recogn

ition

ofpreg

nancy,high

parity

20Lerebo

etal.2015

[31]

Facilitybased

stud

yCross

sectional

415

Preg

nant

wom

enattend

ing

ANCat

governmen

the

alth

facilities

Face

toface

interview

ofpreg

nant

wom

enLate

BookingforANC

Entryin

tocare

after16

weeks

ofge

station

Highparity,un

intend

edpreg

nancy,

perceivedrig

httim

eto

book

ANC

(16weeks

+),no

historyof

abortio

n

21Moh

ammed

and

Berhane2014

[68]

Facilitybased

stud

yCross

sectional

383

Preg

nant

wom

enattend

ing

ANCat

selected

publiche

alth

centres

Face

toface

interview

ofpreg

nant

wom

enTimingof

firstANC

initiation

Entryin

tocare

after16

weeks

ofge

station

Youn

gerage,no

n-attend

ance

ofform

aled

ucation,

incorrectpe

rcep

tionof

timingof

ANC,b

eing

busy

22Tekelaband

Berhanu2014

[32]

Facilitybased

stud

yCross

sectional

401

Preg

nant

wom

enattend

ing

ANCserviceat

governmen

tal

health

centres

Interviewer

administered

structured

questio

nnaire

Late

initiationof

ANC

Entryin

tocare

after16

weeks

ofge

station

Age

(≥25

year),no

n-attend

ance

ofform

aled

ucation,

low

mon

thlyincome,

high

parity,previous

non-useof

ANC,

unintend

edpreg

nancy

From

allthe

iden

tifiedstud

ies,10

06wereexclud

eddu

ringscreen

ingfordu

plication,

and29

53du

ringtitle,abstractan

dfulltext

assessmen

tOne

stud

ywas

exclud

eddu

eto

poor

metho

dologicalq

ualityan

dtherest

21stud

ieswereinclud

edin

themeta-an

alysis

Tesfaye et al. Reproductive Health (2017) 14:150 Page 8 of 17

Employed women were less likely to delay their ANC ascompared to their counterparts. However, the subgroup(delayed ANC initiation ≥12 weeks) (OR, 0.76; 95% CI:0.53, 1.09) showed no association between maternal oc-cupation and delayed booking of ANC. But the overallassociation was not altered. The heterogeneity test

showed an I2 value of 74%, and therefore random effectmodel was considered for the analysis.

Husband’s educationAssociation between husband’s education and delayed initi-ation of ANC was carried out in one group of studies that

Fig. 2 Pooled estimation of delayed initiation of ANC in Ethiopia, 2002–2017. NB (◊: Overall combined pooled proportion, and ■: Originalstudies proportion)

Table 2 Overview of factors associated with delayed initiation of ANC according to the three delay model in Ethiopia, 2002–2017

Category of thefactors

Significantly associated with delayed ANC (COR at 95% CI)

Yes No

Delay one Maternal ageMaternal educationHusband’s educationPregnancy intentionWomen’s autonomyPartner involvementKnowledge on ANCPresence of pregnancy complicationParity

History of abortionMarital status

Delay two Place of residenceMaternal occupationMonthly income

Delay three Means of checking pregnancy Previous ANC utilization

Tesfaye et al. Reproductive Health (2017) 14:150 Page 9 of 17

defined the outcome variable with (≥ 16 weeks), as thisvariable was not reported in the other group of stud-ies. The analysis showed that women having a hus-band who attended formal education were less likelyto delay their first antenatal visit as compared tothose women whose husband had never attended for-mal education (OR, 0.44; 95% CI: 0.23, 0.85). Randomeffect model was implemented for the analysis sincethe I2 value was greater than 50%.

Women’s autonomyWe conducted analysis of a single group of studiesthat defined the outcome variable with (≥12 weeks)due to the unavailability of corresponding data aboutwomen’s autonomy in the other group of studies. Ac-cordingly, the analysis result revealed that women’sautonomy has a significant association with delayedinitiation of ANC (OR, 0.38; 95% CI: 0.15, 0.94). Au-tonomous women were less likely to initiate their firstANC later than non-autonomous women. Randomeffect model was used for the analysis as the I2 testresult is 89%.

Previous use of ANCThe finding revealed that there was no significant associ-ation between previous utilization of ANC and delayedinitiation of ANC (OR, 0.62; 95% CI: 0.34, 1.11). Thiswas the case in both subgroups of the studies, and theoverall analysis result. We assumed a random effectmodel for the analysis as the I2 value 85% showed sub-stantial heterogeneity between the studies.

ParityParity was another predicting factor that affected delayedinitiation of ANC. In this regard, women with no parity(nulliparous) were less likely to have delayed their ANCinitiation as compared to women who were primiparaand above. This was demonstrated in the overall OddsRatio, 0.46 at 95% CI: 0.36, 0.58. There was no differencein the association between parity and delayed initiationof ANC in the subgroup analysis. Since the I2 valuewas 67%, indicating considerable heterogeneity of theincluded studies, we assumed a random effect modelfor the analysis.

Fig. 3 Subgroup and overall association between maternal age (reference category: age 31–49) and delayed initiation of ANC in Ethiopia, 2002–2017

Tesfaye et al. Reproductive Health (2017) 14:150 Page 10 of 17

Partner involvementWe conducted the analysis using studies from both sub-groups and it was found that partner involvement has asignificant association with delayed initiation of ANC.Women who had a partner who was involved in ANCwere less likely to delay thier first ANC initiation com-pared with women with no partner involvement in ANC(OR, 0.24; 95% CI: 0.07, 0.75). We considered a randomeffect model for the analysis because the I2 value was 85%.

Knowledge of ANCThe overall analysis of both groups of studies showedthat knowledge of ANC has association with delayedinitiation of ANC. Knowledgeable women were lesslikely to delay their ANC booking as compared to non-knowledgeable women (OR, 0.40; 95% CI: 0.32, 0.51).Fixed effect model was assumed for the analysis as theChi square test (7.08) with the p-value (0.21) showed

statistically insignificant heterogeneity among the in-cluded studies for this factor analysis.

History of abortionWe found no significant association between history ofabortion and delayed initiation of ANC (OR, 1.16; 95%CI: 0.79, 1.69), and this was true in the analysis result ofboth subgroups of studies. We assumed a random effectmodel since the I2 statistics (77%) showed substantialheterogeneity.

Pregnancy complicationsThere was significant association between the presenceof complications during pregnancy and delayed initiationof ANC on a single group analysis (delayed initiation ofANC ≥ 16 weeks). Women who experienced complica-tions during pregnancy were less likely to delay theirfirst ANC attendance compared to women who did not

Fig. 4 Subgroup and overall association between maternal education (reference category: never attended formal education) and delayedinitiation of ANC in Ethiopia, 2002–2017

Tesfaye et al. Reproductive Health (2017) 14:150 Page 11 of 17

experience complications during pregnancy (OR, 0.23;95% CI: 0.06, 0.95). Random effect model was used forthe analysis since the I2 value was greater than 50%.

Means of identifying pregnancyNo sub-group analysis was performed due to lack of rele-vant statistics with regards to means of identifying preg-nancy in one group (delayed initiation of ANC ≥ 16 weeks).Single group (delayed initiation of ANC ≥ 12 weeks) ana-lysis however showed a significant association betweenmeans of identifying pregnancy with delayed initiation ofANC. Women who identified their pregnancy with aurine test were less likely to delay their first ANC visit ascompared to women who identified their pregnancy usingother means (OR, 0.50; 95% CI: 0.36, 0.69). We consideredrandom effect model for the analysis since the I2 valuewas greater than 50%.

DiscussionMaternal age, maternal education, husband’s education,maternal occupation, place of residence, parity, knowledgeof ANC, women’s autonomy, partner involvement, preg-nancy intention, presence of pregnancy complications,and means of identifying pregnancy were significantly as-sociated factors for delayed initiation of ANC inEthiopia. We found out that nearly two thirds of thewomen in Ethiopia initiated their first ANC late after12th week of pregnancy. Marital status, history ofabortion and previous use of ANC showed no signifi-cant association with delayed initiation of ANC.Timely initiation and continuous attendance of ANC is

believed to improve maternal health outcomes [23, 55].This is the case particularly in developing countries wherethe health status of women is very poor. It is imperative tounderstand the overall level of delayed initiation of ANCand the contributing factors at the country level to inform

Fig. 5 Subgroup and overall association between place of residence (reference category: rural) and delayed initiation of ANC in Ethiopia, 2002–2017

Tesfaye et al. Reproductive Health (2017) 14:150 Page 12 of 17

current efforts to improve maternal outcomes throughadequate utilization of ANC in Ethiopia. The current sys-tematic review supplied a summary of available evidenceon the level of delayed initiation of ANC and associatedfactors in Ethiopia. The importance of systematic reviewsto provide relevant information to transform health caredelivery system and policy modification or ratification waswell documented [56]. This systematic review summarizedup-to-date empirical evidence and fleshed out key areas ofaction regarding delayed initiation of ANC in Ethiopia.This is an important step-forward to ensure maternalhealth program planners and policy makers in the countrymake informed decisions regarding where the correctivemeasures should be instituted and maximized.Even though the WHO [10, 11] recommended initi-

ation of ANC attendance not later than the first trimes-ter of pregnancy, the reviewed evidence showed that themagnitude of delayed initiation of ANC is very high, at64% in Ethiopia. This figure was almost in line with acomparative report of demographic and health survey dataof twenty one sub-Saharan African countries [21] where,on average, more than two-thirds of the reproductive aged

women initiated their first ANC after the first trimester ofpregnancy. This might be due to several socio-cultural,economic and contextual factors including women’s poordecision making power at a household level due to deeplyrooted gender inequality, poor educational status, andpoverty, which in turn could limit the women’s abilityto seek care earlier. The decision to early seek careand assistance during pregnancy among Ethiopianwomen especially in rural areas are linked with many cul-tural practices [57, 58], which were barrier to accessingservices throughout a woman’s pregnancy. Delayed initi-ation of ANC was a significant risk factor for maternaldeath, particularly among the disadvantaged women [22].Hence countries need to prioritise efforts to improve theinitiation of ANC.According to this review, maternal age and education,

husband’s education, parity, knowledge of ANC andwomen’s autonomy were influencing factors for delayedfirst ANC attendance in Ethiopia. The result of the currentreview was in agreement with the systematic review ofstudies [43, 44, 59] conducted in other settings where ma-ternal age, maternal education, husband’s education, and

Fig. 6 Subgroup and overall association between pregnancy intention (reference category: unintended) and delayed initiation of ANC inEthiopia, 2002–2017

Tesfaye et al. Reproductive Health (2017) 14:150 Page 13 of 17

parity were the influencing factors for delayed initiation ofANC. The possible reason for older women aged 31 to 49delaying their first ANC might be that they most likely areuneducated, have poor knowledge of ANC, have experi-enced pregnancies without complications previously, areless fearful unlike younger women and may be more likelyto be multiparous. Education of the mother and husbandcould play a great role in improving awareness of healthmatters in general, and the importance of ANC in particu-lar. Having a better awareness may enable women to seekANC and utilize the service early in pregnancy. This wasparticularly reflected in the systematic review of studiesamong non-western women in industrialized countries[45] where women’s low level of educational status was as-sociated with late entry into ANC.Furthermore, a lack of knowledge about ANC is posi-

tively associated with delayed initiation of ANC. Womenwho had been provided with information regarding ANC,pregnancy risks and danger signs were more likely to initi-ate ANC early compared to women who did not haveknowledge of these issues. This could motivate the womento initiate ANC early to better avoid the risks associatedwith pregnancy. It is anticipated that well informedwomen were more likely to make judicious choices aboutthe proper utilization of ANC. It was also found that

women’s autonomy was a significant predictor of delayedinitiation of ANC where non-autonomous women weremore likely to postpone ANC, which could be due to thefact that they were under the influence of their partner orfamily (especially in male headed households), restrictedto comply with family norms, had lack of family or so-cial support, and a partner who was not available orwho refused to accompany them. This was demon-strated in a systematic review of studies in the develop-ing countries [44], where social support from familymembers, extent of ties within social networks, andobtaining health information from these sources highlyinfluence timely utilization of ANC.Additionally, the meta-analysis revealed that place of

residence, maternal occupation, monthly income, andpartner involvement were significantly associated withdelayed ANC initiation. Rural women were more likelyto delay their first ANC attendance than urban women.This could be explained by the fact that urban womenwould most likely have easy access to health care facilities,have a good awareness of health matters, and have betterexposure to media. Moreover, unemployed women weremore likely to delay initiation of ANC as compared toemployed women. A similar finding was reported in othersystematic reviews [43, 45] where not being in employment

Table 3 Summary of the test statistics of association between the remaining delay one, two and three factors with delayedinitiation of ANC in Ethiopia, 2002–2017

Predictor variable Subgroup OR (95%CI) I2 Combined OR (95% CI) Overall I2

Monthly income [≤1000ETB(50USD)] I 1.77(1.16, 2.72) 65% 2.06(1.23, 3.45) 91%

II 2.26(0.96, 5.29) 94%

Marital status [In marriage] I 0.92(0.55, 1.54) 77% 0.81(0.56, 1.16) 68%

II 0.68(0.38, 1.22) 57%

Maternal occupation [Employed] I 0.76(0.53, 1.09) 83% 0.75(0.61, 0.93) 74%

II 0.74(0.57, 0.97) 66%

Husband education [Attended primary/above] II 0.44(0.23, 0.85) 80%

Women’s autonomy [Autonomous] I 0.38(0.15, 0.94) 89%

Previous use of ANC [Previous use] I 0.65(0.42, 1.02) 45% 0.62(0.34, 1.11) 85%

II 0.53(0.17, 1.67) 92%

Parity [Nulliparity] I 0.51(0.42, 0.61) 0% 0.46(0.36, 0.58) 67%

II 0.42(0.27, 0.66) 81%

Partner involvement [Involved] I 0.44(0.21, 0.91) 85% 0.24(0.07, 0.75) 85%

II 0.14(0.08, 0.22) 85%

Knowledge of ANC [Knowledgeable] I 0.32(0.22, 0.46) 4% 0.40(0.32, 0.51) 29%

II 0.46(0.35, 0.62) 54%

History of abortion [Have history of abortion] I 1.19(0.85, 1.66) 0% 1.16(0.79, 1.69) 77%

II 1.14(0.67, 1.95) 77%

Pregnancy complication [Presence of complication] II 0.23(0.06, 0.95) 97%

Means of identifying pregnancy [Urine] I 0.50(0.36, 0.69) 67%

Subgroup: I = Studies that defined delayed initiation of ANC (after 12 weeks of gestation), Subgroup: II = Studies that defined delayed initiation of ANC (after16 weeks of gestation), I2 is the percentage of total variance due to between study heterogeneity

Tesfaye et al. Reproductive Health (2017) 14:150 Page 14 of 17

explained women’s delayed entry into the care. It was alsoevidenced that women with high economic status weremore likely to receive ANC earlier than those with a lowereconomic status [43, 44]. These financial constraints are inturn related to other barriers to seeking help, includingtransportation costs, the cost of obtaining care, or labora-tory tests [60, 61].Moreover, our finding suggests that women whose

partner was involved in ANC were less likely to delaytheir first ANC attendance than women whose partnerwas not involved in ANC. Partner involvement in termsof initiating and/or supporting the idea to utilize ANCearly, or by accompanying the pregnant mother to thehealth facility may have an important impact on theearly attendance of ANC. In many traditions, the in-volvement of men in reproductive health has not beenconsidered an important issue. In general male partnersdid not accompany their wives to attend ANC and othermaternal health services [43]. The husband’s lack of in-volvement in ANC may immensely affect the women’scapability to initiate ANC early. It was found in a sys-tematic review [62] that the involvement of men in ANChas a positive influence on the overall uptake of the ser-vice and its early attendance.Furthermore, the meta-analysis identified factors such

as pregnancy intention, presence of pregnancy complica-tions and means of identifying pregnancy as an import-ant factors that affect delayed initiation of ANC. Thisfinding is consistent with a systematic review of smallscale studies conducted in both developed and develop-ing countries [63] on the relationship between preg-nancy intention and timely initiation as well as obtainingadequate ANC. It was revealed that unintended preg-nancy has a strong association with delayed initiation offirst ANC services. Another systematic review [44] con-firmed that women whose pregnancy was unintendedtended to initiate ANC later than the first trimester ofpregnancy. With regards to complications during preg-nancy, the findings of this study is similar to a systematicreview of literature [43] conducted in developing coun-tries where pregnant women who did not experience ob-stetric complications were more likely to delay their firstANC compared to their counterparts.In the current systematic review and meta-analysis, we

observed some discrepancies in the included studies indefining the outcome variable “delayed initiation ofANC”. Half of the included studies defined delayed initi-ation of ANC based on the cut-off point of 12 weeks ofgestation, whereas the rest of the studies defined it basedon 16 weeks. However the WHO [10, 11] defined lateANC initiation as entry into care after 12th week of preg-nancy. Conversely, in this review we noticed a contrast-ing type of definition across several studies [27–42] aswell as ANC practice in health facilities [53], implying

that there was poor compliance of the WHO recom-mendation on the timing of first ANC initiation inEthiopia. Countries might prefer to adapt or contextual-ise the original clinical practice guidelines with somechanges, depending on their setting, to effectively imple-ment the recommendations. Even if recommendationsfrom the parent clinical practice guidelines can beadapted, how they are implemented needs to addresslocal issues. Thus countries may need to contextualiseguideline by addressing those implementation issues sothat care becomes more relevant to the local environ-ments [64]. However, at least there should not be incon-sistencies between the implemented specific healthrecommendation within the country’s health care deliv-ery system and the health research arena. Hence, we rec-ommend to concerned parties in the health sector inEthiopia, particularly the health research scholars, thatthere is a need to adhere to the WHO recommendedguideline on the timing of ANC initiation. Moreover,any further adapted or contextualised guideline on thetiming of ANC initiation needs to be followed or imple-mented consistently in a standardized way.The current systematic review and meta-analysis was not

without limitations. The first limitation was the exclusionof qualitative studies from the review, which might revealother important factors affecting women’s behaviour todelay ANC attendance or might otherwise corroborate thequantitative findings. Secondly, since our meta-analysisused Crude Odds Ratios, it might be difficult to fully ascer-tain the effect of the exposure factors on the outcome ofinterest. Thirdly, as all the included studies were cross-sectional by design, it is difficult to establish temporal rela-tionship between the outcome and exposure variables.Lastly, conducting meta-analysis despite the inherentheterogeneity between the included studies might have af-fected the quantitative findings. Our systematic review andmeta-analysis also has some strengths. In this regard, weconsidered selection and inclusion of both published andunpublished literature which has the potential to minimizepublication bias. Moreover, our search strategy was exten-sive using a number of major medical databases andother search engines. Lastly, we conducted a sub-groupanalysis of studies that employed different definitions ofdelayed initiation of ANC to appreciate the independ-ent subgroup findings.

ConclusionThe current review revealed that nearly two thirds ofwomen were delaying their first ANC visit in Ethiopia.The review pointed out various factors attributed to highlevel of delayed initiation of ANC in Ethiopia. Amongthese maternal age, place of residence, maternal educa-tion, husband’s education, maternal occupation, familymonthly income, pregnancy intention, parity, knowledge

Tesfaye et al. Reproductive Health (2017) 14:150 Page 15 of 17

of ANC, women’s autonomy, partner involvement, prob-lem during pregnancy, and means of identifying pregnancyshowed significant association with delayed initiation ofANC. Therefore, intervention efforts to improve ANCutilization in Ethiopia require targeting these impedingfactors. Moreover, strategies should be designed to inten-sify advocacy of female education, women’s empowermentactivities need to be continued through economic re-forms, family planning programs should be strength-ened to reduce unintended pregnancies, and partnerinvolvement in ANC should be promoted through dif-ferent means of communication. Further qualitativestudies are recommended to gain further insight intothe societal and health system barriers that contributeto delayed initiation of ANC in Ethiopia.

Additional files

Additional file 1: PRISMA-P (Preferred Reporting Items for Systematicreview and Meta-Analysis Protocols) 2015 checklist: recommended itemsto address in a systematic review protocol*. (DOC 85 kb)

Additional file 2: Search Strategy. (DOCX 28 kb)

AbbreviationsANC: Antenatal care; CI: Confidence interval; DHS: Demographic and HealthSurvey; ETB: Ethiopian birr; JBI: Joanna Briggs Institute; MAStARI: Meta-Analysisof Statistics Assessment and Review Instrument; M-H: Mantel–Haenszel;NSW: New South Wales; OR: Odds ratio; PRISMA: Preferred Reporting Items forSystematic Reviews and Meta-Analyses Protocols; USD: United States Dollar;WHO: World Health Organization

AcknowledgmentsWe are very grateful to the University of Newcastle, Australia for providing afull scholarship for the corresponding author, and free access to the digitalonline library to search the electronic databases that were considered forthis review. We also would like to acknowledge Haramaya University forproviding free internet access and an office.

Ethical approval and consent to participateNot applicable.

FundingNot applicable.

Availability of data and materialsThe data that support the review findings of this study are available uponsubmitting a reasonable request to the corresponding author.

Authors’ contributionsGT, DL, CC, RS conceptualized the design of the systematic review. GTdrafted the manuscript and is the guarantor of the review. All authorscontributed to the development of the article search strategy, setting studyselection criteria, the strategy for assessment of risk of bias, and dataabstraction form. GT and AS involved in the screening, assessment ofeligibility, selection of studies and critical appraisal as well as data extraction.DL, CC, AS and RS have participated in critically revising the manuscript forimportant intellectual contents. All authors read, provided feedback andapproved the final manuscript.

Consent for publicationNot applicable.

Competing interestsThe authors declare that they have no competing interests.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims in publishedmaps and institutional affiliations.

Author details1Research Centre for Generational Health and Ageing, Faculty of Health andMedicine, University of Newcastle, Newcastle, Australia. 2School of PublicHealth, College of Health and Medical Sciences, Haramaya University, Harar,Ethiopia. 3Mothers and Babies Research Centre, Faculty of Health andMedicine, University of Newcastle, Newcastle, Australia.

Received: 18 July 2017 Accepted: 8 November 2017

References1. Prata N, Passano P, Sreenivas A, Gerdts C. Maternal mortality in developing

countries: challenges in scaling-up priority interventions. Women's Health.2010;6(2):311–27.

2. WHO. Maternal mortality Fact sheet. 2016.3. WHO. Trends in Maternal Mortality: 1990 to 2015 estimates by WHO,

UNICEF, UNFPA, World Bank Group and the United Nations PopulationDivision Excutive summary. 2015.

4. WHO. Trends in Maternal Mortality: 1990 to 2013. 2014.5. Birmeta K, Dibaba Y, Woldeyohannes D. Determinants of maternal health care

utilization in Holeta town, central Ethiopia. BMC Health Serv Res. 2013;13(256):10.6. Ronsmans C, Graham W. Maternal mortality: who, when, where, and why.

Lancet. 2006;368(9542):1189–200.7. Tey N, Lai S. Correlates of and barriers to the utilization of health Services

for Delivery in South Asia and sub-Saharan Africa. ScientificWorldJournal.2013;2013

8. Godefay H, Byass P, Kinsman J, Mulugeta A. Understanding maternalmortality from top-down and bottom-up perspectives: case of Tigrayregion, Ethiopia. J Glob Health. 2015;5(1)

9. FMOH. HSTP Health Sector Transformation Plan. 2015.10. Villar J, P B. WHO antenatal care randomized trial: Manual for the

implementation for the new model. 2002.11. WHO. WHO recommendations on antenatal care for a positive pregnancy

experience. 2016.12. CSA. Ethiopia Demographic and Health Survey 2011 Demographic

surveillance Addis Ababa, Ethiopia and Calverton,Maryland. USA: CSA; 2011.13. CSA. Ethiopian Demographci and Health Survey 2016.14. CSA. Ethiopia Demographic and Health Survey. 2005.15. Ayele D, Belayihun B, Teji K, Ayana D. Factors affecting utilization of

maternal health Care Services in Kombolcha District, eastern Hararghe zone,Oromia regional state, eastern Ethiopia. Int Sch Res Notices. 2014;2014:7.

16. Abosse Z, Woldie M, Ololo S. Factors influencing antenatal care serviceutilization in Hadiya zone. Ethiop J Health Sci. 2010;20(2):75–82.

17. Dutamo Z, Assefa N, Egata G. Maternal health care use among marriedwomen in Hossaina, Ethiopia. BMC Health Serv Res. 2015;15(365):1–9.

18. CSA. Population Projection of Ethiopia for All Regions At Wereda Level from2014–2017. 2014.

19. Melaku Y, Weldearegawi B, Tesfay F, Abera S, Abraham L, Aregay A, et al.Poor linkages in maternal health care services? Evidence on antenatal careand institutional delivery from a community-based longitudinal study inTigray region, Ethiopia. BMC Pregnancy and Childbirth. 2014;14(418)

20. CSA. The Ethiopian mini health and demographic survey. Addis Ababa,Ethiopia, Calverton, Maryland, USA: Central statistical agency of Ethiopia; 2014.

21. Wang W, Alva S, Wang S, Fort A. Level and trends in the use of maternalhealth services in developing countries: DHS comparative reports 26: USAID.2011.

22. Cantwell R, Clutton-Brock T, Cooper G. Saving mothers’ lives: reviewingmaternal deaths to make motherhood safer: 2006-2008. The eighth reportof the confidential enquiries into maternal deaths in the United Kingdom.BJOG. 2011;118(1):149-57.

23. Ziyo F, Matly F, Mehemd G, Dofany E. Relation between prenatal care andpregnancy outcome at Benghazi. Sudanese Journal of public health. 2012;4(4):403–10.

24. Thaddeus S, Maine D. Too far to walk: maternal mortality in context. Soc SciMed. 1994;38(8):1091–110.

25. Berhane Y, Gossaye Y, Emmelinb M, Hogberg U. Women’s health in a ruralsetting in societal transition in Ethiopia. Soc Sci Med. 2001;2001(53):1525–39.

Tesfaye et al. Reproductive Health (2017) 14:150 Page 16 of 17

26. Mekonnen Y, Mekonnen A. Utilization of maternal health Care Services inEthiopia. Calverton: Maryland, USA; 2002.

27. Abosse Z, Woldie M, Ololo S. Factors influencing antenatal care serviceutilization in hadiya zone. Ethiop J Health Sci. 2010;20(2):75–82.

28. Damme T, Workineh D, Gmariam A. Time of Antenatal Care Booking andAssociated Factors Among Pregnant Women Attending Ambo Town HealthFacilities, Central Ethiopia. 2015.

29. Ewnetu W, Assegid S, Wondafrash W, Ewnetu H. Factors associated with lateantenatal care initiation in an Ethiopian clinic. Research. 2015.

30. Gudayu T, Woldeyohannes S, Abdo A. Timing and factors associated withfirst antenatal care booking among pregnant mothers in Gondar town;north West Ethiopia. BMC Pregnancy Childbirth. 2014;14(1):287.

31. Lerebo W, Kidanu A, Tsadik M. Magnitude and associated factors of latebooking for antenatal Care in Public Health Centers of Adigrat town, Tigray.Ethiopia Clinics Mother Child Health. 2015;12(171):2.

32. Tekelab T, Berhanu B. Factors associated with late initiation of antenatal careamong pregnant women attending antenatal Clinic at Public HealthCenters in Kembata Tembaro zone, southern Ethiopia. Science, Technologyand Arts Research Journal. 2014;3(1):108–15.

33. Yilala Y. Assesment of late initiation of antenatal care and associated factorsamong antenatal care attendees in selected health centers of Addis Ababa,Ethiopia (Master Thesis): AAU; 2015.

34. Belayneh T, Adefris M, Andargie G. Previous Early Antenatal ServiceUtilization Improves Timely Booking: Cross-Sectional Study at University ofGondar Hospital, Northwest Ethiopia. 2014.

35. Abuka T, Alemu A, Birhanu B. Assessment Of Timing Of First Antenatal CareBooking And Associated Factors Among Pregnant Women Who AttendAntenatal Care At Health Facilities In Dilla Town, Gedeo Zone, SouthernNations, Nationalities, And Peoples Region, Ethiopia, 2014: AAU; 2014.

36. Bayou Y, Mashalla Y, Tshweneagae G. The adequacy of antenatal careservices among slum residents in Addis Ababa, Ethiopia. BMC PregnancyChildbirth. 2016;16(1):142.

37. Gudayu T. Proportion and factors associated with late antenatal carebooking among pregnant mothers in Gondar town, north West Ethiopia.Afr J Reprod Health. 2015;19(2):93–9.

38. Girum T. Assessment of timing of first antenatal care visit and associatedfactors among pregnant women attending antenatal Care in Dilla TownGovernmental Health Institutions, southern Ethiopia. Alternative &Integrative Medicine. 2016;5(3)

39. Amentie M, Abera M, Abdulahi M. Utilization of antenatal care services andinfluencing factors among women of child bearing age in Assosa District,Benishangul Gumuz regional state, West Ethiopia. Global Journal of MedicalResearch. 2015;15(2)

40. Gebremeskel F, Dibaba Y, Admassu B. Timing of first antenatal careattendance and associated factors among pregnant women in Arba Minchtown and Arba Minch District, Gamo Gofa Zone, South Ethiopia. 2015.

41. Haileselassie D. Magnitude and factors affecting late initiation of antenatalcare service utilization among pregnant women in government healthinstitutions, Mekelle Town, Ethiopia (Master Thesis): AAU; 2010.

42. Hussen S, Melese E, Dembelu M. Timely Initiation of First Antenatal CareVisit of Pregnant Women Attending Antenatal Care Service. 2016.

43. Simkhada B, Teijlingen E, Porter M, Simkhada P. Factors affecting theutilization of antenatal care in developing countries: systematic review ofthe literature. J Adv Nurs. 2007;61(3):244–60.

44. Hajizadeh S, Tehrani F, Simbar M, Farzadfar F. Factors influencing the use ofprenatal care: a systematic review. J midwifery. Reprod Health. 2016;4(1):544–57.

45. Boerleider A, Wiegers T, Manniën J, Francke A, Devillé W. Factors affectingthe use of prenatal care by non-western women in industrialized westerncountries: a systematic review. BMC Pregnancy Childbirth. 2013;13(81)

46. Moher D, Shamseer L, Clarke M, Ghersi D, Liberati A, Petticrew M, et al.Preferred reporting items for systematic review and meta-analysis protocols(PRISMA-P) 2015 statement. Systematic Reviews. 2015;4(1)

47. Moher D, Liberati A, Tetzlaff J, Altman D. Preferred reporting items for systematicreviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7)

48. Reuters T, inventorEndNote X 7.3.1 (BId 8614), Cite While You Write TMPatented technology U.S patent number 8,092,24119888–2015.

49. Institute TJB. Joanna Briggs institute Reviewers' manual 2014 edition.Adelaide: Adelaide University; 2014.

50. Cochrane C, inventorReview Manager (RevMan) [Computer program].Version 5.3. Copenhagen: The Nordic Cochrane Centre, The CochraneCollaboration 2014.

51. Higgins J, Hompson S. Quantifying heterogeneity in a meta-analysis. 2002.52. StatsDirect Ltd. StatsDirect statistical software. England: StatsDirect Ltd.;

2013. http://www.statsdirect.com.53. FMOH. Antenatal Care Module: 13. In: Providing focused antenatal care; 2010.54. Beller E, Glasziou P, Altman D, Hopewell S, Bastian H, Chalmers I, et al.

PRISMA for abstracts: reporting systematic reviews in journal andconference abstracts. PLoS Med. 2013;10(4)

55. Vintzileos A, Ananth C, Smulian J, Scorza W, Knuppel R. The impact of prenatalcare on neonatal deaths in the presence and absence of antenatal high-riskconditions. Am J Obstet Gynecol. 2002;186(5):1011–6.

56. Moat K, Lavis J, Wilson M, Røttingen J, Barnighausen T. Twelve myths aboutsystematic reviews for health system policymaking rebutted. J Health ServRes Policy. 2013;18(1):44–50.

57. USAID/MCHIP. Cultural Barriers to Seeking Maternal Health Care in Ethiopia:A Review of the Literature. 2012.

58. Warren C. Care seeking for maternal health: challenges remain for poorwomen. Ethiopian Journal of Health Development. 2010;24(Special 1):100–4.

59. Jong E, Jansen D, Baarveld F, Schans C, Schellevis F, Reijneveld S.Determinants of late and/or inadequate use of prenatal healthcare in high-income countries: a systematic review. Eur J Pub Health. 2011;22(6):904–13.

60. Yang Y, Yoshitoku Y, Harun R, Junichi S. Factors Affecting the utilization ofantenatal care services among women in Kham district, Xiengkhouangprovince, Lao PDR. Nagoya J Med Sci. 2010;72:23–33.

61. Anyait A, Mukanga D, Oundo G, Nuwaha F. Predictors for health facilitydelivery in Busia district of Uganda: a cross sectional study. BMC PregnancyChildbirth. 2012;12(132)

62. Ayebare E, Mwebaza E, Mwizerwa J, Namutebi E, Kinengyere A, Smyth R.Interventions for male involvement in pregnancy and labour: a systematicreview. 2006.

63. Dibaba Y, Fantahun M, Hindin M. The effects of pregnancy intention on theuse of antenatal care services: systematic review and meta-analysis. ReprodHealth. 2014;10(50)

64. Dizon J, Machingaidze S, Grimmer K. To adopt, to adapt, or tocontextualise? The big question in clinical practice guideline development.BMC Res Notes. 2016;9(442)

65. Zegeye A, Bitew B, Koye D. Prevalence and determinants of early antenatalcare visit among pregnant women attending antenatal Care in Debre BerhanHealth Institutions, Central Ethiopia. Afr J Reprod Health. 2013;17(4):130–6.

66. Fisseha G, Miruts G, Tekie M, W/Michael A, Yemane D, Gerezigiher T.Predictors of timing of first antenatal care booking at public health centersin Mekelle City, northern Ethiopia. J Gynecol Obst. 2015;3(3):55–60.

67. Gulema H, Berhane Y. Timing of first antenatal care visit and its associatedfactors among pregnant women attending public health facilities in AddisAbaba, Ethiopia. Ethiopian Journal of Health Sciences. 2017;27(2):139–46.

68. Mohammed Z, Berhane E. Assessment of timing of first antenatal care (ANC)initiation and associated factors among pregnant women in selected publichealth centres in Addis Ababa, Ethiopia, 2014: AAU; 2014.

• We accept pre-submission inquiries

• Our selector tool helps you to find the most relevant journal

• We provide round the clock customer support

• Convenient online submission

• Thorough peer review

• Inclusion in PubMed and all major indexing services

• Maximum visibility for your research

Submit your manuscript atwww.biomedcentral.com/submit

Submit your next manuscript to BioMed Central and we will help you at every step:

Tesfaye et al. Reproductive Health (2017) 14:150 Page 17 of 17